I ran into a situation some time ago (on a classic 70) which seemed to
contradict
common sense.  A batch job was taking more CPU time (job statistics in
the end of the standard list) when several copies were running together.

A vendor indicated that this can be explained by short and long (disk?)
waits being
included or excluded from the total (don't remember which).  I think the
long disk
waits are not included in the total for the job.  Instead the long disk
waits are
included as CPU time used in overhead.

So adding more memory may appear to increase the CPU time used by a process
when in fact the CPU time is just moved from overhead to job statistics.

"Craig M. Lalley" <[log in to unmask]> wrote in message
news:a5es2508da@enews2.newsguy.com...
> OK Guys,
>
> Here's a fun one.
>
> I have a customer running a 997/800 with 10Gb of memory.
>
> The system is spending on average 9% of the CPU on memory management.
>
> Caaack!
>
> My recommendation is to upgrade to 16GB and max out the configuration.
>
> But wait, HP says NO.  "Adding more memory will have a ?rebound? effect
> and make matters worse."
>
> So am I missing something or HP smoking something?
>
>
> Craig M. Lalley
>
> Systems Performance Specialist
>
> Lund Performance Solutions
>
> Phone (541) 812-7600 | Email [log in to unmask]
>
> Fax (425) 920-4995| Web www.lund.com
>
> NOTICE: This communication may contain privileged or other confidential
> information. If you are not the intended recipient or believe that you
> may have received this communication in error, please reply to the
> sender indicating that fact and delete the copy you received. In
> addition, you should not print, copy, retransmit, disseminate, or
> otherwise use the information. Thank you.
>
> * To join/leave the list, search archives, change list settings, *
> * etc., please visit http://raven.utc.edu/archives/hp3000-l.html *
>

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *